Real-Time Driver Monitoring Systems through Modality and View Analysis
This addresses driver safety by improving efficiency in monitoring systems, but it is incremental as it builds on existing methods by optimizing for latency.
The paper tackled the problem of driver distraction detection by developing real-time algorithms that neglect temporal relations in video frames, achieving 97.5% AUC-PR with reduced computation compared to video-based models, and found the top view with infrared channel to be most informative.
Driver distractions are known to be the dominant cause of road accidents. While monitoring systems can detect non-driving-related activities and facilitate reducing the risks, they must be accurate and efficient to be applicable. Unfortunately, state-of-the-art methods prioritize accuracy while ignoring latency because they leverage cross-view and multimodal videos in which consecutive frames are highly similar. Thus, in this paper, we pursue time-effective detection models by neglecting the temporal relation between video frames and investigate the importance of each sensing modality in detecting drives' activities. Experiments demonstrate that 1) our proposed algorithms are real-time and can achieve similar performances (97.5\% AUC-PR) with significantly reduced computation compared with video-based models; 2) the top view with the infrared channel is more informative than any other single modality. Furthermore, we enhance the DAD dataset by manually annotating its test set to enable multiclassification. We also thoroughly analyze the influence of visual sensor types and their placements on the prediction of each class. The code and the new labels will be released.